Devils Lake is a saline lake in a large, closed drainage basin in northeastern North Dakota. Previous studies determined that major-ion and nutrient concentrations in Devils Lake are strongly affected by microbially mediated sulfate reduction and dissolution of sulfate and carbonate minerals in the bottom sediments. These studies documented substantial spatial variability in the magnitude of calculated benthic fluxes coincident with the horizontal salinity gradient in Devils Lake. The purpose of the present study is to evaluate seasonal variability in benthic-flux rates, and to understand the effect of these fluxes on the major-ion and nutrient chemistries in Devils Lake between May and October 1991.

During the study period, the water column was well mixed, and specific conductance, pH, and temperature did not vary with depth. Dissolved oxygen was enriched near the lake surface due to photosynthesis. Major-ion concentrations and nutrient concentrations did not vary with depth. Because the water-quality data were obtained during open-water periods, the vertical profiles reflect well-mixed conditions. However, the first and last profiles for the study period did document near-bottom maxima of major cations. Secchi-disk depth varied from 0.82 meter on May 7,1991, to 2.13 meters on June 5, 1991. The mean Secchi-disk depth during the study period was 1.24 meters. Seasonal variations in Secchi-disk depths were attributed to variations in primary productivity and phytoplankton communities.

Nutrient cycles in Devils Lake were evaluated using gross primary productivity rate data, sediment trap data, and major-ion and nutrient benthic-flux rate data. Gross primary productivity rate was smallest in May (0.076 gram of carbon per square meter per day) and largest in September (1.8 grams of carbon per square meter per day). Average gross primary productivity for the study period was 0.87 gram of carbon per square meter per day. Average gross primary productivity is consistent with historic data from Devils Lake and with data from other eutrophic lakes.

The average flux of organic carbon for the study period was 12 grams per square meter per day. The calculated carbon to nitrogen to phosphorus ratio (317:25:1) is similar to the Redfield ratio (106:16:1); therefore, most organic matter probably is derived from lacustrine phytoplankton.

Calculated benthic-flux rates indicated that bottom sediments are important sources of majorions and nutrients to Devils Lake. Only one of the cores collected during this study indicated a net sulfate flux from the lake into the sediments. Seasonal variations in major-ion and nutrient benthic fluxes generally were small. However, there were important differences between the calculated benthic fluxes for this study and the calculated benthic fluxes for 1990. Calculated benthic fluxes of bicarbonate, ammonia, and phosphorus for this study were smaller than calculated benthic fluxes for 1990. The large differences between fluxes for 1990 and 1991 were attributed to calm, stratified water-column conditions in 1990 and well-mixed water-column conditions in 1991.

The role of benthic fluxes in the chemical mass balances in Devils Lake was evaluated by calculating response times for major ions and nutrients in Devils Lake. The calculated response times for major ions in Devils Lake ranged from 6.7 years for bicarbonate to 34 years for sulfur (as 804). The response times for major ions are significantly shorter than previous estimates that did not include benthic fluxes. In addition, the relatively short response times for nitrogen (4.2 years) and phosphorus (0.95 year) indicate that nutrients are recycled rapidly between bottom sediments and the lake. During the study period, benthic fluxes were the dominant source of major ions and nutrients to Devils Lake and greatly reduced the response times of all major ions and nutrients for Devils Lake. As a result, bottom-sediment processes appear to buffer major-ion and nutrient concentrations in the lake. Any future attempt to evaluate water quality in Devils Lake should include the effects of bottom-sediment processes.